Why I Start With a 4:1 UNUN for Non-Harmonic Multiband Wires

The design question arrives as soon as the wanted bands are not harmonically related. A wire that is useful on 60, 40 and 30 metres, for example, does not present the same resistance and reactance three times. My answer is not to make the transformer pretend otherwise. I measure the installed load, use a moderate transformation when it helps the tuner, and define the common-mode boundary separately.

Non-Harmonic Bands Do Not Repeat the Same Load

An end-fed half-wave can present another high-voltage, high-impedance point near harmonically related resonances. A wire chosen for unrelated bands does not offer that repetition. Its electrical length, current distribution and terminal impedance change independently on each band.

That is not a defect. It is simply a multiband feed-system problem. A 13–14 m sloper or flat-top may be an attractive physical span for 60, 40 and 30 metres, but its installed feedpoint impedance depends on exact length, conductor, height, shape, soil, surroundings, return conductor and feedline participation. A free-space length table cannot replace the sweep.

Measure resistance and reactance—R + jX—at the transformer plane with the intended geometry connected. A tuner has to deal with both parts. Quoting only the resistance, or only the lowest SWR point, hides much of the load the transformer and tuner actually see.

What a Nominal 4:1 Ratio Actually Does

For an ideal transformer, the impedance ratio is the square of the turns ratio:

A 4:1 impedance transformation can ideally map a 200 Ω resistive load toward 50 Ω. That example explains the attraction of the ratio; it does not say that every non-harmonic wire is 200 Ω, or that the transformation direction and ratio remain ideal across HF.

Real transmission-line and magnetic transformers have finite magnetising impedance, leakage, winding and enclosure capacitance, conductor loss, core loss and frequency-dependent mode behaviour. With a reactive antenna load, the transformed result is also reactive. Voltage, current, waveform, duty cycle and temperature decide whether the implementation remains inside its operating domain.

This is why I say “start with 4:1,” not “4:1 is always correct.” When the measured loads cluster in a domain where moderate transformation reduces the extremes presented to the tuner, it is a sensible candidate. When the loads do not, change the wire, return geometry, feedline, ratio or matching topology.

Why I Do Not Choose 9:1 or 49:1 by Habit

A nominal 9:1 ratio ideally maps 450 Ω toward 50 Ω. A nominal 49:1 ratio ideally maps 2,450 Ω toward 50 Ω. Those arithmetic examples assume resistive loads and ideal behaviour. They are not band assignments.

A higher ratio can be useful when the measured load calls for it, but it also changes winding voltage and current, parasitic sensitivity, loaded bandwidth and thermal conditions. Selecting 49:1 because the radiator is “end-fed,” or 9:1 because the impedance is “random,” replaces measurement with a label.

The comparison must use the same installed antenna and declared planes. Measure the load without hiding line transformation, characterize each candidate under representative complex loads, and compare the complete path at equal accepted power. A lower transmitter SWR can coexist with greater transformer or feedline loss.

The Counterpoise Sets the Return Conditions

An UNUN transforms the impedance between two terminals. It does not create the second terminal's current path. The intentional counterpoise, coax exterior, mast, bonding network, soil coupling and station wiring can all participate in the return.

Counterpoise wires around 5.3 m, 3.5 m or 2.6 m can be legitimate trial geometries for particular 60, 40 or 30 m installations. They are not universal stabilizers. Their electrical lengths, routes, heights and coupling change current division, terminal impedance, common-mode current and pattern on every band.

Commission the return structure rather than merely attaching it. Measure current on each intended conductor and at several locations on the coax exterior. Repeat the impedance sweep after a controlled counterpoise change, then restore the baseline. A repeatable change tells you the return path matters; it does not yet prove whether loss or radiation improved.

The Separate Choke Is Deliberate

I prefer to keep common-mode control independent of impedance transformation. The UNUN is selected for the measured differential load. A separately specified current choke is placed where it should stop an unintended continuation of the coax-exterior return path.

The choke need not be at one universal distance. A deliberate section of coax exterior before the choke can form part of an end-fed return structure. In another installation, the useful boundary may be close to the transformer. Map the current and choose the boundary; do not infer it from a fraction of the internal coax velocity factor.

A current balun is not disqualified by this preference. For a genuinely balanced installed two-terminal load, a properly designed current balun can provide transformation and common-mode impedance. It must still be verified under the actual complex loads, currents, voltages and thermal conditions. The reason I often use the separate pair is that real amateur antenna installations frequently do not preserve the balance suggested by the drawing.

The Tuner Finishes a Match, Not the Proof

A tuner can transform the load appearing at its output into a load acceptable to the transmitter. A moderate transformer can keep that load inside a more comfortable part of the tuner's range. Neither fact guarantees low loss.

For each band, record tuner settings, accepted power and component temperature. Watch for high-voltage and high-current operating points. If a small change in feedline length or counterpoise geometry makes the tuner move dramatically, the complete current system is telling you that the load domain is not stable.

The 14 m wire is therefore a useful candidate, not a universal “sweet spot.” Longer options such as a roughly 24.5 m sloper or 35.5 m Inverted-L create different electrical lengths, current distributions, loads and patterns. They may be excellent at one site and awkward at another. Choose among them with the intended bands, available geometry, tuner map and radiated pattern included.

Bandwidth and Pattern Need Their Own Evidence

Operating away from a self-resonant point does not automatically create wide useful bandwidth. The transmitter-side match can be broad because the transformer, line and tuner hide rapid changes elsewhere, or because loss damps the impedance variation. Define bandwidth by the limits that matter: match, loss, voltage, current, temperature, common-mode current and pattern.

Likewise, avoiding one resonance does not avoid deep pattern nulls. A wire that is several half wavelengths long can develop multiple current regions, lobes and nulls. Height, slope, bends, return-current geometry and nearby conductors change the installed result. Model the complete current geometry and verify important directions with controlled field or on-air comparisons.

My Commissioning Sequence

• Define the operating bands: include the frequencies, directions, elevation angles, power and duty cycle that matter.
• Document the physical system: radiator, return conductors, feedline route, supports, ground and nearby metal.
• Measure R+jX: calibrate at the transformer plane and sweep the complete installed antenna on every intended band.
• Compare candidate transformations: use representative complex loads and measure insertion loss, voltage, current and temperature.
• Map common-mode current: inspect the coax exterior before and after adding or moving the separate choke.
• Confirm tuner margin: record repeatable settings and check loss and thermal behaviour at the intended operating condition.
• Verify the radiated result: use an installed-geometry model and calibrated or time-controlled A/B/A field measurements at equal accepted power.
• Complete the safety assessment: include RF exposure, high-voltage wire ends, transformer heating, mechanical loading, lightning protection and overhead-line clearance.

Questions, antenna-factor records or height trials to share? Contact RF.Guru.